Skill-Building Exercise 2 Solutions

Required Setup

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.5.1     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.1
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
## 
## Attaching package: 'scales'
## 
## 
## The following object is masked from 'package:purrr':
## 
##     discard
## 
## 
## The following object is masked from 'package:readr':
## 
##     col_factor

## [[1]]
##  [1] "lubridate" "forcats"   "stringr"   "dplyr"     "purrr"     "readr"    
##  [7] "tidyr"     "tibble"    "ggplot2"   "tidyverse" "stats"     "graphics" 
## [13] "grDevices" "utils"     "datasets"  "methods"   "base"     
## 
## [[2]]
##  [1] "gt"        "lubridate" "forcats"   "stringr"   "dplyr"     "purrr"    
##  [7] "readr"     "tidyr"     "tibble"    "ggplot2"   "tidyverse" "stats"    
## [13] "graphics"  "grDevices" "utils"     "datasets"  "methods"   "base"     
## 
## [[3]]
##  [1] "scales"    "gt"        "lubridate" "forcats"   "stringr"   "dplyr"    
##  [7] "purrr"     "readr"     "tidyr"     "tibble"    "ggplot2"   "tidyverse"
## [13] "stats"     "graphics"  "grDevices" "utils"     "datasets"  "methods"  
## [19] "base"     
## 
## [[4]]
##  [1] "fst"       "scales"    "gt"        "lubridate" "forcats"   "stringr"  
##  [7] "dplyr"     "purrr"     "readr"     "tidyr"     "tibble"    "ggplot2"  
## [13] "tidyverse" "stats"     "graphics"  "grDevices" "utils"     "datasets" 
## [19] "methods"   "base"     
## 
## [[5]]
##  [1] "ggridges"  "fst"       "scales"    "gt"        "lubridate" "forcats"  
##  [7] "stringr"   "dplyr"     "purrr"     "readr"     "tidyr"     "tibble"   
## [13] "ggplot2"   "tidyverse" "stats"     "graphics"  "grDevices" "utils"    
## [19] "datasets"  "methods"   "base"     
## 
## [[6]]
##  [1] "ggrepel"   "ggridges"  "fst"       "scales"    "gt"        "lubridate"
##  [7] "forcats"   "stringr"   "dplyr"     "purrr"     "readr"     "tidyr"    
## [13] "tibble"    "ggplot2"   "tidyverse" "stats"     "graphics"  "grDevices"
## [19] "utils"     "datasets"  "methods"   "base"     
## 
## [[7]]
##  [1] "patchwork" "ggrepel"   "ggridges"  "fst"       "scales"    "gt"       
##  [7] "lubridate" "forcats"   "stringr"   "dplyr"     "purrr"     "readr"    
## [13] "tidyr"     "tibble"    "ggplot2"   "tidyverse" "stats"     "graphics" 
## [19] "grDevices" "utils"     "datasets"  "methods"   "base"

Loading data

gss <- read_fst("gss2022.fst")

Task 1: Abortion Attitudes in American Society

a. Data Preparation

First, let’s examine the abortion variable (abany) and the other variables requested in the guidelines:

print("Distribution of abortion views:")

## [1] "Distribution of abortion views:"

table(gss$abany, useNA = "ifany")

## 
##                           yes                            no 
##                         16626                         22628 
##                    don't know                           iap 
##                             0                             0 
##            I don't have a job                   dk, na, iap 
##                             0                             0 
##                     no answer    not imputable_(2147483637) 
##                             0                             0 
##    not imputable_(2147483638)                       refused 
##                             0                             0 
##                skipped on web                    uncodeable 
##                             0                             0 
## not available in this release    not available in this year 
##                             0                             0 
##                  see codebook                          <NA> 
##                             0                         33136

print("\nDistribution of political views:")

## [1] "\nDistribution of political views:"

table(gss$polviews, useNA = "ifany")

## 
##             extremely liberal                       liberal 
##                          2081                          7623 
##              slightly liberal  moderate, middle of the road 
##                          7900                         23992 
##         slightly conservative                  conservative 
##                          9596                          9361 
##        extremely conservative                    don't know 
##                          2165                             0 
##                           iap            I don't have a job 
##                             0                             0 
##                   dk, na, iap                     no answer 
##                             0                             0 
##    not imputable_(2147483637)    not imputable_(2147483638) 
##                             0                             0 
##                       refused                skipped on web 
##                             0                             0 
##                    uncodeable not available in this release 
##                             0                             0 
##    not available in this year                  see codebook 
##                             0                             0 
##                          <NA> 
##                          9672

print("\nDistribution of education:")

## [1] "\nDistribution of education:"

table(gss$degree, useNA = "ifany")

## 
##         less than high school                   high school 
##                         14192                         36446 
##      associate/junior college                    bachelor's 
##                          4355                         11248 
##                      graduate                    don't know 
##                          5953                             0 
##                           iap            I don't have a job 
##                             0                             0 
##                   dk, na, iap                     no answer 
##                             0                             0 
##    not imputable_(2147483637)    not imputable_(2147483638) 
##                             0                             0 
##                       refused                skipped on web 
##                             0                             0 
##                    uncodeable not available in this release 
##                             0                             0 
##    not available in this year                  see codebook 
##                             0                             0 
##                          <NA> 
##                           196

print("\nDistribution of gender:")

## [1] "\nDistribution of gender:"

table(gss$sex, useNA = "ifany")

## 
##                          male                        female 
##                         31977                         40301 
##                    don't know                           iap 
##                             0                             0 
##            I don't have a job                   dk, na, iap 
##                             0                             0 
##                     no answer    not imputable_(2147483637) 
##                             0                             0 
##    not imputable_(2147483638)                       refused 
##                             0                             0 
##                skipped on web                    uncodeable 
##                             0                             0 
## not available in this release    not available in this year 
##                             0                             0 
##                  see codebook                          <NA> 
##                             0                           112

Now, let’s clean the data according to the assignment requirements.

gss <- gss %>%
  mutate(
    abortion_view = case_when(
      abany == "yes" ~ "Support",
      abany == "no" ~ "Oppose",
      TRUE ~ NA_character_  
    ),
    abortion_view = factor(abortion_view),
    
    # Clean political views - three categories
    political_view = case_when(
      polviews %in% c("extremely liberal", "liberal", "slightly liberal") ~ "Liberal",
      polviews == "moderate, middle of the road" ~ "Moderate",
      polviews %in% c("extremely conservative", "conservative", "slightly conservative") ~ "Conservative",
      TRUE ~ NA_character_
    ),
    political_view = factor(political_view, 
                          levels = c("Liberal", "Moderate", "Conservative")),
    
    education = case_when(
      degree %in% c("less than high school", "high school") ~ "High School or Less",
      degree == "associate/junior college" ~ "Some College",
      degree %in% c("bachelor's", "graduate") ~ "Bachelor's or Higher",
      TRUE ~ NA_character_
    ),
    education = factor(education, 
                     levels = c("High School or Less", "Some College", "Bachelor's or Higher")),
    
    # Clean gender - binary as coded in GSS
    gender = case_when(
      sex == "male" ~ "Male",
      sex == "female" ~ "Female",
      TRUE ~ NA_character_
    ),
    gender = factor(gender)
  )

Importantly, let’s check our recoding:

print("Cleaned abortion views:")

## [1] "Cleaned abortion views:"

table(gss$abortion_view, useNA = "ifany")

## 
##  Oppose Support    <NA> 
##   22628   16626   33136

print("\nCleaned political views:")

## [1] "\nCleaned political views:"

table(gss$political_view, useNA = "ifany")

## 
##      Liberal     Moderate Conservative         <NA> 
##        17604        23992        21122         9672

print("\nCleaned education:")

## [1] "\nCleaned education:"

table(gss$education, useNA = "ifany")

## 
##  High School or Less         Some College Bachelor's or Higher 
##                50638                 4355                17201 
##                 <NA> 
##                  196

print("\nCleaned gender:")

## [1] "\nCleaned gender:"

table(gss$gender, useNA = "ifany")

## 
## Female   Male   <NA> 
##  40301  31977    112

b. Table Creation

Let’s create a combined table showing support for abortion rights across all the social/political characteristics (note, you can do it all at once, but we will do it in separate chunks similar to the tutorial):

political_summary <- gss %>%
  filter(!is.na(abortion_view), !is.na(political_view)) %>%
  group_by(political_view) %>%
  summarize(
    n = n(),
    n_support = sum(abortion_view == "Support"),
    pct_support = round(100 * n_support/n, 1)
  )
political_summary

## # A tibble: 3 × 4
##   political_view     n n_support pct_support
##   <fct>          <int>     <int>       <dbl>
## 1 Liberal        10395      6189        59.5
## 2 Moderate       14040      5881        41.9
## 3 Conservative   12450      3854        31

You can even do a check by taking the Number supporting reported (e.g., 6189) divided by the sample size for that given political views label (e.g., 10395) = 59.5.

education_summary <- gss %>%
  filter(!is.na(abortion_view), !is.na(education)) %>%
  group_by(education) %>%
  summarize(
    n = n(),
    n_support = sum(abortion_view == "Support"),
    pct_support = round(100 * n_support/n, 1)
  )
education_summary

## # A tibble: 3 × 4
##   education                n n_support pct_support
##   <fct>                <int>     <int>       <dbl>
## 1 High School or Less  27601     10180        36.9
## 2 Some College          2373      1083        45.6
## 3 Bachelor's or Higher  9193      5333        58

gender_summary <- gss %>%
  filter(!is.na(abortion_view), !is.na(gender)) %>%
  group_by(gender) %>%
  summarize(
    n = n(),
    n_support = sum(abortion_view == "Support"),
    pct_support = round(100 * n_support/n, 1)
  )
gender_summary

## # A tibble: 2 × 4
##   gender     n n_support pct_support
##   <fct>  <int>     <int>       <dbl>
## 1 Female 22000      9142        41.6
## 2 Male   17229      7475        43.4

Now we have all the values we want to report. Now, we can combine all the values with a bind_rows. There are other ways to complete this part of the assignment, including calculating all of the values at once and then putting them into the formatted gt table.

all_values <- bind_rows(
  political_summary %>% 
    mutate(characteristic = "Political Views",
           category = political_view) %>%
    select(-political_view),
  
  education_summary %>%
    mutate(characteristic = "Education",
           category = education) %>%
    select(-education),
  
  gender_summary %>%
    mutate(characteristic = "Gender",
           category = gender) %>%
    select(-gender)
)

all_values %>%
  gt(
    groupname_col = "characteristic"
  ) %>%
  tab_header(
    title = md("**Support for Legal Abortion by Social Characteristics**"),
    subtitle = "GSS, 1977-2022"
  ) %>%
  fmt_number(
    columns = c(n),
    decimals = 0,
    use_seps = TRUE
  ) %>%
  cols_label(
    category = "Category",
    n = "Sample Size",
    n_support = "Number Supporting",
    pct_support = "% Support"
  ) %>%
  tab_style(
    style = cell_text(weight = "bold"),
    locations = cells_row_groups()
  )

Support for Legal Abortion by Social Characteristics
GSS, 1977-2022
Sample Size	Number Supporting	% Support	Category
Political Views
10,395	6189	59.5	Liberal
14,040	5881	41.9	Moderate
12,450	3854	31.0	Conservative
Education
27,601	10180	36.9	High School or Less
2,373	1083	45.6	Some College
9,193	5333	58.0	Bachelor's or Higher
Gender
22,000	9142	41.6	Female
17,229	7475	43.4	Male

Make sure to always double check what displays in your formatted table vs. what you calculated to ensure it matches.

c. Visualization

Now, let’s create a visualization showing opposition to abortion over time for all three variables as requested in the guidelines. The steps are very similar but instead we are doing for “Oppose” instead of “Support”. That could have been handled all at once, but we are breaking down the assignment per task.

political_trends <- gss %>%
  filter(!is.na(abortion_view), !is.na(political_view)) %>%
  group_by(year, political_view) %>%
  summarize(
    n = n(),
    prop_oppose = mean(abortion_view == "Oppose"),
    .groups = "drop"
  )

education_trends <- gss %>%
  filter(!is.na(abortion_view), !is.na(education)) %>%
  group_by(year, education) %>%
  summarize(
    n = n(),
    prop_oppose = mean(abortion_view == "Oppose"),
    .groups = "drop"
  )

gender_trends <- gss %>%
  filter(!is.na(abortion_view), !is.na(gender)) %>%
  group_by(year, gender) %>%
  summarize(
    n = n(),
    prop_oppose = mean(abortion_view == "Oppose"),
    .groups = "drop"
  )

all_trends <- bind_rows(
  political_trends %>%
    mutate(characteristic = "Political Views",
           category = political_view),
  
  education_trends %>%
    mutate(characteristic = "Education",
           category = education),
  
  gender_trends %>%
    mutate(characteristic = "Gender",
           category = gender)
)

last_points <- all_trends %>%
  group_by(characteristic, category) %>%
  slice_max(order_by = year, n = 1)

ggplot(all_trends, 
       aes(x = year, y = prop_oppose, color = category)) +
  geom_line(size = 1.2) +
  scale_color_manual(
    values = c(
      "Liberal" = "#2E74C0",
      "Moderate" = "#6C6C6C",
      "Conservative" = "#CB454A",
      "High School or Less" = "#6C3C89",
      "Some College" = "#C098D0",
      "Bachelor's or Higher" = "#005F73",
      "Female" = "#E69F00",
      "Male" = "#56B4E9"
    )
  ) +
  geom_label_repel(
    data = last_points,
    aes(label = category),
    nudge_x = 2,
    direction = "y",
    hjust = 0,
    segment.size = 0.3,
    segment.color = "grey70",
    box.padding = 0.5,
    point.padding = 0.5,
    size = 3.5,
    fontface = "bold",
    label.size = 0.1,
    label.r = unit(0.2, "lines"),
    fill = alpha("white", 0.7)
  ) +
  facet_wrap(~characteristic, scales = "free_y") +
  scale_y_continuous(
    labels = scales::percent_format(),
    limits = c(0, 1.0),
    breaks = seq(0, 1.0, by = 0.2),
    expand = c(0.01, 0.01)
  ) +
  scale_x_continuous(
    breaks = seq(1975, 2025, by = 10),
    limits = c(1972, 2027),
    expand = c(0.01, 0.01)
  ) +
  labs(
    title = "Opposition to Legal Abortion Across Social Groups",
    subtitle = "Percentage opposing abortion when 'woman wants it for any reason'",
    x = NULL,
    y = "Proportion Opposing",
    caption = "Source: General Social Survey, 1977-2022"
  ) +
  theme_minimal() +
  theme(
    plot.title = element_text(
      face = "bold",
      size = 14,
      margin = margin(b = 10)
    ),
    plot.subtitle = element_text(
      color = "grey40",
      size = 12,
      margin = margin(b = 20)
    ),
    axis.line = element_line(color = "black", size = 0.5),
    axis.ticks = element_line(color = "black", size = 0.5),
    panel.grid = element_blank(),
    strip.text = element_text(face = "bold", size = 12),
    strip.background = element_rect(fill = "grey95", color = NA),
    legend.position = "none",
    axis.text = element_text(size = 9, color = "black"),
    plot.margin = margin(20, 20, 20, 20)
  )

Not required but for ease of interpretation:

milestone_years <- c(1977, 1985, 1990, 2000, 2010, 2022)

opposition_milestones <- all_trends %>%
  filter(year %in% milestone_years) %>%
  mutate(pct_oppose = round(prop_oppose * 100, 1)) %>%
  select(year, characteristic, category, pct_oppose) %>%
  pivot_wider(
    id_cols = c(characteristic, category),
    names_from = year,
    values_from = pct_oppose
  ) %>%
  arrange(characteristic, category)

opposition_milestones

## # A tibble: 8 × 8
##   characteristic  category             `1977` `1985` `1990` `2000` `2010` `2022`
##   <chr>           <fct>                 <dbl>  <dbl>  <dbl>  <dbl>  <dbl>  <dbl>
## 1 Education       High School or Less    65.6   67.3   61.2   63.6   62.2   49.5
## 2 Education       Some College           56.2   50     40.4   63.9   59.8   45.5
## 3 Education       Bachelor's or Higher   42.8   46.7   43.7   48.9   43.4   29.7
## 4 Gender          Female                 63.3   64.5   56.7   60.2   57.9   41.3
## 5 Gender          Male                   61.1   61.3   56.4   60.1   54.5   45.2
## 6 Political Views Liberal                55.1   48.5   41.6   44.7   37.6   18  
## 7 Political Views Moderate               62.2   65.6   54.5   58.3   57     41.8
## 8 Political Views Conservative           66.8   69.3   68.9   72.5   70.4   67.8

d. Interpretation

Based on the table and visualization, I’ll provide a clear and focused interpretation of the abortion attitude patterns:

The GSS data reveal significant differences and substantial changes in abortion attitudes across social groups from 1977 to 2022, with three distinct patterns emerging.

Political ideology shows the most dramatic divergence. Conservative opposition to abortion has remained consistently high (67-70% in recent decades) with minimal change from the 66.8% opposition recorded in 1977. In stark contrast, liberal opposition has plummeted from 55.1% in 1977 to just 18.0% in 2022. This growing polarization is particularly pronounced after 1990, with the ideological gap expanding from approximately 20 percentage points to over 50 percentage points by 2022. Moderates follow a middle trajectory, declining from 62.2% opposition in 1977 to 41.8% in 2022.

Educational differences show persistent but narrowing gaps. Opposition among those with bachelor’s degrees or higher has declined modestly from 42.8% in 1977 to 29.7% in 2022. Those with high school education or less demonstrated a larger decline from 65.6% to 49.5% during the same period. The “some college” group typically falls between these extremes, showing more fluctuation over time, with notable variation in 2000 (63.9%) and 2022 (45.5%). note: this is directly related to the smaller sample size relative to any of the other sub-categories (i.e., smaller sample sizes will be more prone to year-to-year fluctuation since they are more sensitive to outliers and to compositional differences).

Gender differences remain surprisingly modest despite abortion directly related to women healthcare issues (there are theories about direct interests, which here are not really supported). Both women and men show parallel declines in opposition, from about 63.3% and 61.1% respectively in 1977 to 41.3% and 45.2% in 2022. This relatively small gender gap (never exceeding 4 percentage points in recent surveys) suggests that abortion attitudes are not significantly structured by gender.

These patterns demonstrate that abortion views have become increasingly defined by political ideology and educational attainment rather than gender, reflecting abortion’s transformation to a defining partisan divide in American politics.

Task 2: LGBTQ+ Rights in European Context

ess <- read_fst("All-ESS-Data.fst")

a. Data Preparation

First, let’s examine the variables we’ll be working with:

print("LGBTQ+ rights variable (freehms):")

## [1] "LGBTQ+ rights variable (freehms):"

table(ess$freehms, useNA = "ifany")

## 
##      1      2      3      4      5      7      8      9 
## 152529 167624  69465  43607  36444   1592  18458    836

print("\nEducation variable (eisced):")

## [1] "\nEducation variable (eisced):"

table(ess$eisced, useNA = "ifany")

## 
##     0     1     2     3     4     5     6     7    55    77    88    99 
## 73306 38823 71917 74258 87348 49268 42651 49558  1052   483   497  1394

print("\nDomicile variable (domicil):")

## [1] "\nDomicile variable (domicil):"

table(ess$domicil, useNA = "ifany")

## 
##      1      2      3      4      5      7      8      9 
## 107787  55216 149256 148750  27534    137    493   1382

print("\nCountry distribution:")

## [1] "\nCountry distribution:"

table(ess$cntry, useNA = "ifany")

## 
##    AL    AT    BE    BG    CH    CY    CZ    DE    DK    EE    ES    FI    FR 
##  1201 15225 17451 13240 16925  6065 20090 34425 12408 16856 19452 19532 19038 
##    GB    GR    HR    HU    IE    IL    IS    IT    LT    LU    LV    ME    MK 
## 20979 12558  6535 16642 22233 16218  3975 10178 11652  3187  3921  2478  1429 
##    NL    NO    PL    PT    RO    RS    RU    SE    SI    SK    TR    UA    XK 
## 18329 16065 17689 17881  2146  3548 12458 18216 13484 11292  4272  9987  1295

Now, let’s clean the data according to the assignment requirements (need also to consult survey documentation/ESS data portal):

ess_clean <- ess %>%
  filter(cntry %in% c("FR", "HU")) %>%
  mutate(
    lgbtq_rights = case_when(
      freehms %in% c(1, 2) ~ "Support",
      freehms == 3 ~ "Neutral",
      freehms %in% c(4, 5) ~ "Oppose",
      TRUE ~ NA_character_
    ),
    lgbtq_rights = factor(lgbtq_rights, 
                         levels = c("Support", "Neutral", "Oppose")),
    
    education = case_when(
      eisced %in% 1:2 ~ "Lower Secondary or Less",
      eisced %in% 3:4 ~ "Upper Secondary",
      eisced %in% 5:7 ~ "Tertiary",
      TRUE ~ NA_character_
    ),
    education = factor(education, 
                      levels = c("Lower Secondary or Less",
                               "Upper Secondary", 
                               "Tertiary")),
    
    residence = case_when(
      domicil == 1 ~ "Major city",              
      domicil %in% c(2, 3) ~ "Urban periphery", 
      domicil %in% c(4, 5) ~ "Rural",           
      TRUE ~ NA_character_
    ),
    residence = factor(residence, 
                      levels = c("Major city", "Urban periphery", "Rural")),
    
    country = case_when(
      cntry == "FR" ~ "France",
      cntry == "HU" ~ "Hungary",
      TRUE ~ NA_character_
    ),
    country = factor(country)
  )

Let’s check what we did:

print("Checking LGBTQ+ rights distribution:")

## [1] "Checking LGBTQ+ rights distribution:"

table(ess_clean$lgbtq_rights, useNA = "ifany")

## 
## Support Neutral  Oppose    <NA> 
##   22504    5343    6157    1676

print("\nChecking education distribution:")

## [1] "\nChecking education distribution:"

table(ess_clean$education, useNA = "ifany")

## 
## Lower Secondary or Less         Upper Secondary                Tertiary 
##                    8355                   15609                    8278 
##                    <NA> 
##                    3438

print("\nChecking residence distribution:")

## [1] "\nChecking residence distribution:"

table(ess_clean$residence, useNA = "ifany")

## 
##      Major city Urban periphery           Rural            <NA> 
##            7780           15078           12801              21

print("\nChecking country distribution:")

## [1] "\nChecking country distribution:"

table(ess_clean$country, useNA = "ifany")

## 
##  France Hungary 
##   19038   16642

b. Table Creation

Now let’s create the four required tables, starting with France and education:

france_edu <- ess_clean %>%
  filter(country == "France", 
         !is.na(lgbtq_rights), 
         !is.na(education)) %>%
  mutate(total_n = n()) %>%
  group_by(education) %>%
  summarize(
    n = n(),
    pct_of_sample = round(100 * n/first(total_n), 1),
    pct_oppose = round(100 * mean(lgbtq_rights == "Oppose"), 1)
  )
france_edu

## # A tibble: 3 × 4
##   education                   n pct_of_sample pct_oppose
##   <fct>                   <int>         <dbl>      <dbl>
## 1 Lower Secondary or Less  4168          26.7       13.9
## 2 Upper Secondary          6490          41.6        6.6
## 3 Tertiary                 4928          31.6        3.9

france_edu_table <- france_edu %>%
  gt() %>%
  tab_header(
    title = md("**Opposition to LGBTQ+ Rights by Education in France**"),
    subtitle = "ESS 2002-2020"
  ) %>%
  fmt_number(
    columns = n,
    decimals = 0,
    use_seps = TRUE
  ) %>%
  fmt_number(
    columns = c(pct_of_sample, pct_oppose),
    decimals = 1
  ) %>%
  cols_label(
    education = "Education Level",
    n = "Group Size",
    pct_of_sample = "% of Sample",
    pct_oppose = "% Opposing"
  ) %>%
  tab_source_note(
    source_note = md("Data: European Social Survey. *% opposing represents conditional probabilities.*")
  )

france_edu_table

Opposition to LGBTQ+ Rights by Education in France
ESS 2002-2020
Education Level	Group Size	% of Sample	% Opposing
Lower Secondary or Less	4,168	26.7	13.9
Upper Secondary	6,490	41.6	6.6
Tertiary	4,928	31.6	3.9
Data: European Social Survey. % opposing represents conditional probabilities.

Repeating, but for Hungary now:

hungary_edu <- ess_clean %>%
  filter(country == "Hungary", 
         !is.na(lgbtq_rights), 
         !is.na(education)) %>%
  mutate(total_n = n()) %>%
  group_by(education) %>%
  summarize(
    n = n(),
    pct_of_sample = round(100 * n/first(total_n), 1),
    pct_oppose = round(100 * mean(lgbtq_rights == "Oppose"), 1)
  )
hungary_edu

## # A tibble: 3 × 4
##   education                   n pct_of_sample pct_oppose
##   <fct>                   <int>         <dbl>      <dbl>
## 1 Lower Secondary or Less  3476          23.1       38.7
## 2 Upper Secondary          8434          56.1       29.4
## 3 Tertiary                 3126          20.8       22.8

hungary_edu_table <- hungary_edu %>%
  gt() %>%
  tab_header(
    title = md("**Opposition to LGBTQ+ Rights by Education in Hungary**"),
    subtitle = "ESS 2002-2020"
  ) %>%
  fmt_number(
    columns = n,
    decimals = 0,
    use_seps = TRUE
  ) %>%
  fmt_number(
    columns = c(pct_of_sample, pct_oppose),
    decimals = 1
  ) %>%
  cols_label(
    education = "Education Level",
    n = "Group Size",
    pct_of_sample = "% of Sample",
    pct_oppose = "% Opposing"
  ) %>%
  tab_source_note(
    source_note = md("Data: European Social Survey. *% opposing represents conditional probabilities.*")
  )

hungary_edu_table

Opposition to LGBTQ+ Rights by Education in Hungary
ESS 2002-2020
Education Level	Group Size	% of Sample	% Opposing
Lower Secondary or Less	3,476	23.1	38.7
Upper Secondary	8,434	56.1	29.4
Tertiary	3,126	20.8	22.8
Data: European Social Survey. % opposing represents conditional probabilities.

Now France with place of residence:

france_res <- ess_clean %>%
  filter(country == "France", 
         !is.na(lgbtq_rights), 
         !is.na(residence)) %>%
  mutate(total_n = n()) %>%
  group_by(residence) %>%
  summarize(
    n = n(),
    pct_of_sample = round(100 * n/first(total_n), 1),
    pct_oppose = round(100 * mean(lgbtq_rights == "Oppose"), 1)
  )
france_res

## # A tibble: 3 × 4
##   residence           n pct_of_sample pct_oppose
##   <fct>           <int>         <dbl>      <dbl>
## 1 Major city       3556          18.8        8.8
## 2 Urban periphery  8543          45.3        8.5
## 3 Rural            6780          35.9        8.3

france_res_table <- france_res %>%
  gt() %>%
  tab_header(
    title = md("**Opposition to LGBTQ+ Rights by Residence in France**"),
    subtitle = "ESS 2002-2020"
  ) %>%
  fmt_number(
    columns = n,
    decimals = 0,
    use_seps = TRUE
  ) %>%
  fmt_number(
    columns = c(pct_of_sample, pct_oppose),
    decimals = 1
  ) %>%
  cols_label(
    residence = "Place of Residence",
    n = "Group Size",
    pct_of_sample = "% of Sample",
    pct_oppose = "% Opposing"
  ) %>%
  tab_source_note(
    source_note = md("Data: European Social Survey. *% opposing represents conditional probabilities.*")
  )

france_res_table

Opposition to LGBTQ+ Rights by Residence in France
ESS 2002-2020
Place of Residence	Group Size	% of Sample	% Opposing
Major city	3,556	18.8	8.8
Urban periphery	8,543	45.3	8.5
Rural	6,780	35.9	8.3
Data: European Social Survey. % opposing represents conditional probabilities.

Repeating for Hungary:

hungary_res <- ess_clean %>%
  filter(country == "Hungary", 
         !is.na(lgbtq_rights), 
         !is.na(residence)) %>%
  mutate(total_n = n()) %>%
  group_by(residence) %>%
  summarize(
    n = n(),
    pct_of_sample = round(100 * n/first(total_n), 1),
    pct_oppose = round(100 * mean(lgbtq_rights == "Oppose"), 1)
  )
hungary_res

## # A tibble: 3 × 4
##   residence           n pct_of_sample pct_oppose
##   <fct>           <int>         <dbl>      <dbl>
## 1 Major city       3870          25.6       27.5
## 2 Urban periphery  5921          39.2       30.5
## 3 Rural            5315          35.2       31.7

hungary_res_table <- hungary_res %>%
  gt() %>%
  tab_header(
    title = md("**Opposition to LGBTQ+ Rights by Residence in Hungary**"),
    subtitle = "ESS 2002-2020"
  ) %>%
  fmt_number(
    columns = n,
    decimals = 0,
    use_seps = TRUE
  ) %>%
  fmt_number(
    columns = c(pct_of_sample, pct_oppose),
    decimals = 1
  ) %>%
  cols_label(
    residence = "Place of Residence",
    n = "Group Size",
    pct_of_sample = "% of Sample",
    pct_oppose = "% Opposing"
  ) %>%
  tab_source_note(
    source_note = md("Data: European Social Survey. *% opposing represents conditional probabilities.*")
  )

hungary_res_table

Opposition to LGBTQ+ Rights by Residence in Hungary
ESS 2002-2020
Place of Residence	Group Size	% of Sample	% Opposing
Major city	3,870	25.6	27.5
Urban periphery	5,921	39.2	30.5
Rural	5,315	35.2	31.7
Data: European Social Survey. % opposing represents conditional probabilities.

c. Visualization

Now let’s create the four required visualizations following the conditional probability pattern, notably shown in Tutorial 4. First France x Edu:

ggplot(france_edu, 
       aes(x = pct_oppose, y = reorder(education, pct_oppose))) +

    geom_col(
        width = 0.7,
        fill = "#2E74C0",
        alpha = 0.9
    ) +

    geom_point(
        size = 3,
        color = "#1B4B78"
    ) +

    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#2E74C0"
    ) +

    scale_x_continuous(
        limits = c(0, max(france_edu$pct_oppose) * 1.2),
        breaks = seq(0, 15, by = 5),
        labels = scales::label_number(suffix = "%")
    ) +

    labs(
        title = "Opposition to LGBTQ+ Rights by Education in France",
        subtitle = "ESS 2002-2020",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 14),
        plot.subtitle = element_text(color = "grey40", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(30, 30, 30, 30)
    )

Now, Hungary x Edu:

ggplot(hungary_edu, 
       aes(x = pct_oppose, y = reorder(education, pct_oppose))) +

    geom_col(
        width = 0.7,
        fill = "#CB454A",
        alpha = 0.9
    ) +

    geom_point(
        size = 3,
        color = "#7A2A2D"
    ) +

    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#CB454A"
    ) +

    scale_x_continuous(
        limits = c(0, max(hungary_edu$pct_oppose) * 1.2),
        breaks = seq(0, 40, by = 10),
        labels = scales::label_number(suffix = "%")
    ) +

    labs(
        title = "Opposition to LGBTQ+ Rights by Education in Hungary",
        subtitle = "ESS 2002-2020",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 14),
        plot.subtitle = element_text(color = "grey40", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(30, 30, 30, 30)
    )

France x Place:

ggplot(france_res, 
       aes(x = pct_oppose, y = reorder(residence, pct_oppose))) +

    geom_col(
        width = 0.7,
        fill = "#2E74C0",
        alpha = 0.9
    ) +

    geom_point(
        size = 3,
        color = "#1B4B78"
    ) +

    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#2E74C0"
    ) +

    scale_x_continuous(
        limits = c(0, max(france_res$pct_oppose) * 1.2),
        breaks = seq(0, 10, by = 2),
        labels = scales::label_number(suffix = "%")
    ) +

    labs(
        title = "Opposition to LGBTQ+ Rights by Residence in France",
        subtitle = "ESS 2002-2020",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 14),
        plot.subtitle = element_text(color = "grey40", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(30, 30, 30, 30)
    )

Hungary x Place:

ggplot(hungary_res, 
       aes(x = pct_oppose, y = reorder(residence, pct_oppose))) +

    geom_col(
        width = 0.7,
        fill = "#CB454A",
        alpha = 0.9
    ) +

    geom_point(
        size = 3,
        color = "#7A2A2D"
    ) +

    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#CB454A"
    ) +

    scale_x_continuous(
        limits = c(0, max(hungary_res$pct_oppose) * 1.2),
        breaks = seq(0, 35, by = 5),
        labels = scales::label_number(suffix = "%")
    ) +

    labs(
        title = "Opposition to LGBTQ+ Rights by Residence in Hungary",
        subtitle = "ESS 2002-2020",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 14),
        plot.subtitle = element_text(color = "grey40", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(30, 30, 30, 30)
    )

d. Interpretation

The ESS data reveal differences in LGBTQ+ rights attitudes between France and Hungary, with notable patterns across the conditional probabilties for educational and geographic groups.

France demonstrates remarkably low opposition to LGBTQ+ rights, with a clear educational gradient. Among those with lower secondary education or less, 13.9% oppose LGBTQ+ rights, while this drops to just 6.6% among those with upper secondary education and falls further to a mere 3.9% among those with tertiary education. This represents a substantial 10 percentage point difference between the lowest and highest education groups. Interestingly, geography shows minimal impact on attitudes in France, with opposition remaining consistently low regardless of location (8.8% in major cities, 8.5% in urban peripheries, and 8.3% in rural areas).

In stark contrast, Hungary exhibits much higher opposition across all demographics. Educational differences are particularly pronounced, with 38.7% of those with lower secondary education or less opposing LGBTQ+ rights, compared to 29.4% among those with upper secondary education and 22.8% among those with tertiary education. This represents a 15.9 percentage point gap between the lowest and highest education groups - even larger than in France. Unlike France, Hungary also demonstrates geographic variation, with major cities showing lower opposition (27.5%) compared to urban peripheries (30.5%) and rural areas (31.7%).

The cross-national comparison is especially revealing. The opposition level among Hungary’s most progressive demographic (tertiary-educated at 22.8%) exceeds that of France’s least progressive group (lower secondary or less at 13.9%). This indicates fundamental differences in national contexts. While education emerges as important on highlighting patterns of attitudes in both countries, residential differences appear as relatively important only in Hungary.

Bonus

Here is one way to do it (note: the output should be the same as above):

combined_data <- bind_rows(
  france_edu %>% mutate(country = "France", demographic = "Education"),
  hungary_edu %>% mutate(country = "Hungary", demographic = "Education"),
  france_res %>% mutate(country = "France", demographic = "Residence", 
                       education = residence) %>% select(-residence),
  hungary_res %>% mutate(country = "Hungary", demographic = "Residence", 
                        education = residence) %>% select(-residence)
)

combined_table <- combined_data %>%
  pivot_wider(
    names_from = country,
    values_from = c(n, pct_of_sample, pct_oppose)
  ) %>%
  select(demographic, education, 
         n_France, pct_of_sample_France, pct_oppose_France,
         n_Hungary, pct_of_sample_Hungary, pct_oppose_Hungary) %>%
  arrange(demographic, education)

combined_gt <- combined_table %>%
  gt(groupname_col = "demographic") %>%
  tab_header(
    title = md("**Opposition to LGBTQ+ Rights: France vs. Hungary**"),
    subtitle = "Comparison by Education Level and Place of Residence (ESS 2002-2020)"
  ) %>%
  fmt_number(
    columns = c(n_France, n_Hungary),
    decimals = 0,
    use_seps = TRUE
  ) %>%
  fmt_number(
    columns = contains("pct"),
    decimals = 1
  ) %>%
  cols_label(
    education = "Group",
    n_France = "Group Size",
    pct_of_sample_France = "% of Sample",
    pct_oppose_France = "% Opposing",
    n_Hungary = "Group Size",
    pct_of_sample_Hungary = "% of Sample",
    pct_oppose_Hungary = "% Opposing"
  ) %>%
  tab_spanner(
    label = md("**France**"),
    columns = c(n_France, pct_of_sample_France, pct_oppose_France)
  ) %>%
  tab_spanner(
    label = md("**Hungary**"),
    columns = c(n_Hungary, pct_of_sample_Hungary, pct_oppose_Hungary)
  ) %>%
  tab_style(
    style = list(
      cell_fill(color = "aliceblue")
    ),
    locations = cells_body(
      columns = c(n_France, pct_of_sample_France, pct_oppose_France)
    )
  ) %>%
  tab_style(
    style = list(
      cell_fill(color = "#FFEFEF")
    ),
    locations = cells_body(
      columns = c(n_Hungary, pct_of_sample_Hungary, pct_oppose_Hungary)
    )
  ) %>%
  tab_style(
    style = cell_text(weight = "bold"),
    locations = cells_row_groups()
  ) %>%
  tab_source_note(
    source_note = md("Data: European Social Survey. *% opposing represents conditional probabilities.*")
  )
combined_gt 

Opposition to LGBTQ+ Rights: France vs. Hungary
Comparison by Education Level and Place of Residence (ESS 2002-2020)
Group	France			Hungary
	Group Size	% of Sample	% Opposing	Group Size	% of Sample	% Opposing
Education
Lower Secondary or Less	4,168	26.7	13.9	3,476	23.1	38.7
Upper Secondary	6,490	41.6	6.6	8,434	56.1	29.4
Tertiary	4,928	31.6	3.9	3,126	20.8	22.8
Residence
Major city	3,556	18.8	8.8	3,870	25.6	27.5
Urban periphery	8,543	45.3	8.5	5,921	39.2	30.5
Rural	6,780	35.9	8.3	5,315	35.2	31.7
Data: European Social Survey. % opposing represents conditional probabilities.

And two examples of a visualization. One simpler where we just combine all the visuals created before together. And another where we try something else to compare all the values together:

france_edu_plot <- ggplot(france_edu, 
       aes(x = pct_oppose, y = reorder(education, pct_oppose))) +
    geom_col(
        width = 0.7,
        fill = "#2E74C0",
        alpha = 0.9
    ) +
    geom_point(
        size = 3,
        color = "#1B4B78"
    ) +
    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#2E74C0"
    ) +
    scale_x_continuous(
        limits = c(0, max(france_edu$pct_oppose) * 1.2),
        breaks = seq(0, 15, by = 5),
        labels = scales::label_number(suffix = "%")
    ) +
    labs(
        title = "France: Education",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(10, 10, 10, 10)
    )

hungary_edu_plot <- ggplot(hungary_edu, 
       aes(x = pct_oppose, y = reorder(education, pct_oppose))) +
    geom_col(
        width = 0.7,
        fill = "#CB454A",
        alpha = 0.9
    ) +
    geom_point(
        size = 3,
        color = "#7A2A2D"
    ) +
    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#CB454A"
    ) +
    scale_x_continuous(
        limits = c(0, max(hungary_edu$pct_oppose) * 1.2),
        breaks = seq(0, 40, by = 10),
        labels = scales::label_number(suffix = "%")
    ) +
    labs(
        title = "Hungary: Education",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(10, 10, 10, 10)
    )

france_res_plot <- ggplot(france_res, 
       aes(x = pct_oppose, y = reorder(residence, pct_oppose))) +
    geom_col(
        width = 0.7,
        fill = "#2E74C0",
        alpha = 0.9
    ) +
    geom_point(
        size = 3,
        color = "#1B4B78"
    ) +
    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#2E74C0"
    ) +
    scale_x_continuous(
        limits = c(0, max(france_res$pct_oppose) * 1.2),
        breaks = seq(0, 10, by = 2),
        labels = scales::label_number(suffix = "%")
    ) +
    labs(
        title = "France: Residence",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(10, 10, 10, 10)
    )

hungary_res_plot <- ggplot(hungary_res, 
       aes(x = pct_oppose, y = reorder(residence, pct_oppose))) +
    geom_col(
        width = 0.7,
        fill = "#CB454A",
        alpha = 0.9
    ) +
    geom_point(
        size = 3,
        color = "#7A2A2D"
    ) +
    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.5,
        size = 4,
        fontface = "bold",
        color = "#CB454A"
    ) +
    scale_x_continuous(
        limits = c(0, max(hungary_res$pct_oppose) * 1.2),
        breaks = seq(0, 35, by = 5),
        labels = scales::label_number(suffix = "%")
    ) +
    labs(
        title = "Hungary: Residence",
        x = "Percentage Opposing",
        y = NULL
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 12),
        panel.grid.major.x = element_line(color = "grey95", size = 0.3),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank(),
        plot.margin = margin(10, 10, 10, 10)
    )

combined_plots <- (france_edu_plot + hungary_edu_plot) / (france_res_plot + hungary_res_plot)

combined_plots_with_title <- combined_plots + 
  plot_annotation(
    title = "Opposition to LGBTQ+ Rights in France and Hungary",
    subtitle = "ESS 2002-2020",
    caption = "Data: European Social Survey. Percentages represent conditional probabilities.",
    theme = theme(
      plot.title = element_text(face = "bold", size = 16),
      plot.subtitle = element_text(size = 12, color = "grey40"),
      plot.caption = element_text(size = 10, color = "grey40", hjust = 0)
    )
  )

combined_plots_with_title

combined_data <- bind_rows(
  france_edu %>% mutate(country = "France", demographic = "Education", group = education) %>% select(-education),
  hungary_edu %>% mutate(country = "Hungary", demographic = "Education", group = education) %>% select(-education),
  france_res %>% mutate(country = "France", demographic = "Residence", group = residence) %>% select(-residence),
  hungary_res %>% mutate(country = "Hungary", demographic = "Residence", group = residence) %>% select(-residence)
)

combined_data$category <- paste(combined_data$country, combined_data$demographic, combined_data$group)

all_in_one_plot <- ggplot(combined_data, 
       aes(x = reorder(category, pct_oppose), y = pct_oppose, fill = country)) +
    geom_col(
        width = 0.7,
        alpha = 0.9
    ) +
    geom_point(
        aes(color = country),
        size = 3
    ) +
    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        hjust = -0.3,
        size = 3.5,
        fontface = "bold"
    ) +
    scale_y_continuous(
        labels = scales::label_percent(scale = 1),
        limits = c(0, max(combined_data$pct_oppose) * 1.2),
        breaks = seq(0, 40, by = 10)
    ) +
    scale_fill_manual(
        values = c(
            "France" = "#2E74C0",
            "Hungary" = "#CB454A"
        )
    ) +
    scale_color_manual(
        values = c(
            "France" = "#1B4B78",
            "Hungary" = "#7A2A2D"
        )
    ) +
    labs(
        title = "Opposition to LGBTQ+ Rights: All Conditional Probabilities",
        subtitle = "Ordered from highest to lowest opposition (ESS 2002-2020)",
        x = NULL,
        y = "Percentage Opposing",
        caption = "Data: European Social Survey"
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 16, margin = margin(b = 10)),
        plot.subtitle = element_text(color = "grey40", size = 12, margin = margin(b = 20)),
        axis.text.x = element_text(),
        panel.grid.major.x = element_blank(),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_line(color = "grey95", size = 0.3),
        legend.position = "top",
        legend.title = element_blank(),
        legend.text = element_text(size = 10, face = "bold"),
        plot.margin = margin(20, 20, 20, 20)
    ) +
    coord_flip()
    
all_in_one_plot

Trying one more improved version:

improved_categories <- combined_data %>%
  mutate(
    demographic_group = paste(demographic, group),
    country_color = country
  )

improved_plot <- ggplot(improved_categories, 
       aes(x = reorder(demographic_group, pct_oppose), y = pct_oppose, fill = country_color)) +
    geom_col(
        width = 0.7,
        position = position_dodge(width = 0.8),
        alpha = 0.9
    ) +
    geom_point(
        aes(color = country_color),
        position = position_dodge(width = 0.8),
        size = 3
    ) +
    geom_text(
        aes(label = sprintf("%.1f%%", pct_oppose)),
        position = position_dodge(width = 0.8),
        hjust = -0.3,
        size = 3.5,
        fontface = "bold"
    ) +
    scale_y_continuous(
        labels = scales::label_percent(scale = 1),
        limits = c(0, max(improved_categories$pct_oppose) * 1.2),
        breaks = seq(0, 40, by = 10)
    ) +
    scale_fill_manual(
        values = c(
            "France" = "#2E74C0",
            "Hungary" = "#CB454A"
        ),
        name = "Country"
    ) +
    scale_color_manual(
        values = c(
            "France" = "#1B4B78",
            "Hungary" = "#7A2A2D"
        ),
        name = "Country"
    ) +
    labs(
        title = "Opposition to LGBTQ+ Rights by Demographic Categories",
        subtitle = "Comparing France and Hungary (ESS 2002-2020)",
        x = NULL,
        y = "Percentage Opposing",
        caption = "Data: European Social Survey"
    ) +
    theme_minimal() +
    theme(
        plot.title = element_text(face = "bold", size = 16, margin = margin(b = 10)),
        plot.subtitle = element_text(color = "grey40", size = 12, margin = margin(b = 20)),
        axis.text.x = element_text(size = 9),
        axis.text.y = element_text(size = 9),
        panel.grid.major.x = element_blank(),
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_line(color = "grey95", size = 0.3),
        legend.position = "top",
        legend.title = element_text(size = 10, face = "bold"),
        legend.text = element_text(size = 10),
        plot.margin = margin(20, 20, 20, 20)
    ) +
    coord_flip() +
    guides(color = "none")
    
improved_plot